/usr/lib/python2.7/dist-packages/cssutils/tokenize2.py is in python-cssutils 1.0.2-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 | #!/usr/bin/env python
# -*- coding: utf-8 -*-
"""New CSS Tokenizer (a generator)
"""
__all__ = ['Tokenizer', 'CSSProductions']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from cssproductions import *
from helper import normalize
import itertools
import re
import sys
_TOKENIZER_CACHE = {}
class Tokenizer(object):
"""
generates a list of Token tuples:
(Tokenname, value, startline, startcolumn)
"""
_atkeywords = {
u'@font-face': CSSProductions.FONT_FACE_SYM,
u'@import': CSSProductions.IMPORT_SYM,
u'@media': CSSProductions.MEDIA_SYM,
u'@namespace': CSSProductions.NAMESPACE_SYM,
u'@page': CSSProductions.PAGE_SYM,
u'@variables': CSSProductions.VARIABLES_SYM
}
_linesep = u'\n'
unicodesub = re.compile(r'\\[0-9a-fA-F]{1,6}(?:\r\n|[\t\r\n\f\x20])?').sub
cleanstring = re.compile(r'\\((\r\n)|[\n\r\f])').sub
def __init__(self, macros=None, productions=None, doComments=True):
"""
inits tokenizer with given macros and productions which default to
cssutils own macros and productions
"""
if isinstance(macros, dict):
macros_hash_key = sorted(macros.items())
else:
macros_hash_key = macros
hash_key = str((macros_hash_key, productions))
if hash_key in _TOKENIZER_CACHE:
(tokenmatches, commentmatcher, urimatcher) = _TOKENIZER_CACHE[hash_key]
else:
if not macros:
macros = MACROS
if not productions:
productions = PRODUCTIONS
tokenmatches = self._compile_productions(self._expand_macros(macros,
productions))
commentmatcher = [x[1] for x in tokenmatches if x[0] == 'COMMENT'][0]
urimatcher = [x[1] for x in tokenmatches if x[0] == 'URI'][0]
_TOKENIZER_CACHE[hash_key] = (tokenmatches, commentmatcher, urimatcher)
self.tokenmatches = tokenmatches
self.commentmatcher = commentmatcher
self.urimatcher = urimatcher
self._doComments = doComments
self._pushed = []
def _expand_macros(self, macros, productions):
"""returns macro expanded productions, order of productions is kept"""
def macro_value(m):
return '(?:%s)' % macros[m.groupdict()['macro']]
expanded = []
for key, value in productions:
while re.search(r'{[a-zA-Z][a-zA-Z0-9-]*}', value):
value = re.sub(r'{(?P<macro>[a-zA-Z][a-zA-Z0-9-]*)}',
macro_value, value)
expanded.append((key, value))
return expanded
def _compile_productions(self, expanded_productions):
"""compile productions into callable match objects, order is kept"""
compiled = []
for key, value in expanded_productions:
compiled.append((key, re.compile('(?:%s)' % value, re.U).match))
return compiled
def push(self, *tokens):
"""Push back tokens which have been pulled but not processed."""
self._pushed = itertools.chain(tokens, self._pushed)
def clear(self):
self._pushed = []
def tokenize(self, text, fullsheet=False):
"""Generator: Tokenize text and yield tokens, each token is a tuple
of::
(name, value, line, col)
The token value will contain a normal string, meaning CSS unicode
escapes have been resolved to normal characters. The serializer
escapes needed characters back to unicode escapes depending on
the stylesheet target encoding.
text
to be tokenized
fullsheet
if ``True`` appends EOF token as last one and completes incomplete
COMMENT or INVALID (to STRING) tokens
"""
def _repl(m):
"used by unicodesub"
num = int(m.group(0)[1:], 16)
if num <= sys.maxunicode:
return unichr(num)
else:
return m.group(0)
def _normalize(value):
"normalize and do unicodesub"
return normalize(self.unicodesub(_repl, value))
line = col = 1
# The current starting character. We just increase this instead of
# splitting off the beginning of text to increase performance.
pos = 0
# check for BOM first as it should only be max one at the start
(BOM, matcher), productions = self.tokenmatches[0], self.tokenmatches[1:]
match = matcher(text, pos)
if match:
found = match.group(0)
yield (BOM, found, line, col)
pos += len(found)
# check for @charset which is valid only at start of CSS
if has_at(text, pos, '@charset '):
found = '@charset ' # production has trailing S!
yield (CSSProductions.CHARSET_SYM, found, line, col)
pos += len(found)
col += len(found)
# Avoid repeated function call
_len_text = len(text)
_orig_text = text
while pos < _len_text:
# do pushed tokens before new ones
for pushed in self._pushed:
yield pushed
# speed test for most used CHARs, sadly . not possible :(
c = text[pos]
if c in u',:;{}>[]': # + but in num!
yield ('CHAR', c, line, col)
col += 1
pos += 1
else:
# check all other productions, at least CHAR must match
for name, matcher in productions:
# TODO: USE bad comment?
if (fullsheet and name == 'CHAR' and
has_at(text, pos, u'/*')):
# before CHAR production test for incomplete comment
possiblecomment = u'%s*/' % text[pos:]
match = self.commentmatcher(possiblecomment)
if match and self._doComments:
yield ('COMMENT', possiblecomment, line, col)
pos = _len_text # ate all remaining text
break
match = matcher(text, pos) # if no match try next production
if match:
found = match.group(0) # needed later for line/col
# The ident regex also matches the beginning of
# functions, but we can't put the function regex before
# the ident regex, as otherwise 'and(' is recognized as
# function (even though it is valid in media queries).
# So we're doing this: if we find an ident, but the next
# character is a open parenthesis, we instead skip and
# let the FUNCTION production take over - except if the
# ident is "and"
if (name == 'IDENT' and
found.lower() != "and" and
match.end(0) < len(text) and
text[match.end(0)] == '('):
continue
if fullsheet:
# check if found may be completed into a full token
if ('INVALID' == name and
suffix_eq(text, pos, found)):
# complete INVALID to STRING with start char " or '
name, found = 'STRING', '%s%s' % (found, found[0])
elif 'FUNCTION' == name and\
u'url(' == _normalize(found):
# url( is a FUNCTION if incomplete sheet
# FUNCTION production MUST BE after URI production
for end in (u"')", u'")', u')'):
possibleuri = '%s%s' % (text[pos:], end)
match = self.urimatcher(possibleuri)
if match:
name, found = 'URI', match.group(0)
break
if name in ('DIMENSION', 'IDENT', 'STRING', 'URI',
'HASH', 'COMMENT', 'FUNCTION', 'INVALID',
'UNICODE-RANGE'):
# may contain unicode escape, replace with normal
# char but do not _normalize (?)
value = self.unicodesub(_repl, found)
if name in ('STRING', 'INVALID'): #'URI'?
# remove \ followed by nl (so escaped) from string
value = self.cleanstring('', value)
else:
if 'ATKEYWORD' == name:
try:
# get actual ATKEYWORD SYM
name = self._atkeywords[_normalize(found)]
except KeyError, e:
# might also be misplace @charset...
if ('@charset' == found and
has_at(text, pos + len(found), u' ')):
# @charset needs tailing S!
name = CSSProductions.CHARSET_SYM
found += u' '
else:
name = 'ATKEYWORD'
value = found # should not contain unicode escape (?)
if self._doComments or (not self._doComments and
name != 'COMMENT'):
yield (name, value, line, col)
pos += len(found)
nls = found.count(self._linesep)
line += nls
if nls:
col = len(found[found.rfind(self._linesep):])
else:
col += len(found)
break
# Make sure we didn't accidentally modify text in the process
assert text is _orig_text
if fullsheet:
yield ('EOF', u'', line, col)
def has_at(text, pos, string):
"""Check if text has substring string as position pos.
This is effectively equal to ``text[pos:pos+len(string)] == string``.
If ``pos`` is 0, this is equal to ``text.startswith(string)``.
:param str text: The text to check against.
:param int pos: The starting position where the substring should be
searched.
:param str string: The string to search.
"""
return text[pos:pos+len(string)] == string
def suffix_eq(text, pos, expected):
"""Check if the suffix of text starting at pos equals expected.
This is effectively equivalent to ``text[pos:] == expected``, but avoids
allocating a huge substring.
If ``pos`` is 0, this is equal to ``text == expected``.
:param str text: Text to compare.
:param int pos: Starting position for the suffix.
:param str expected: String to compare to.
"""
if not pos:
return text == expected
return (len(text) - pos == len(expected) and
all(text[i + pos] == expected[i] for i in range(len(expected))))
|